New York City is known as “The City That Never Sleeps”. While the excitement of the constant hustle & bustle may be a draw for some, it also can wear thin after a while. With over 8 million people in merely 302 square miles, there are bound to be some ruffled feathers. This projects seeks to understand noise complaints made to 311 from 2012 to 2018. We look for patterns based on time, attributes of populations within zip codes, and the effect of music venues on the complaints made in their local environments.
We created an overall correlation graph to understand the relationships between 311 noise complaints and selected demographic variables over time. The number of housing units and the total population immediately jump out as the most correlated, but interestingly so does the population under the age of 18.
We explored the most correlated variable by zip code to see if there were any neighborhood patterns. We also visualized the correlation between noise complaints and population under 18 at this level to discover geographic variation.
We created a Heat Map to understand how noise complaints to 311 have evolved in NYC from 2012 to 2018, and to reveal any patterns that happen each year. The map displays that there has been a general increase in noise complaint calls to 311 over the 7 year period. Additionally, it looks like there are more noise complaint calls placed to 311 during the warmer months (approximately April to October), and that there are generally more calls placed on the weekends than during the week.
There is a positive relationship between the percent of people in a zip code with Bachelor’s Degrees (or higher) and the number of noise complaints in that zip code per capita. There is a large concentration of zip codes with few Bachelors degrees and few noise complaints, and then as the percentage of people with Bachelor’s degrees in a zipcode increases, the minimum number of noise complaints increases as well. It’s interesting to note that the upper end of the noise complaint range seems to be concentrated towards the lower end and middle of the percent of people with bachelor’s degrees range, not the upper end.
There wasn’t as strong of a relationship between median income and noise complaints as there was between some of the other census variables and noise complaints, so we decided to look at the relationship more closely to learn about what was going on. We isolated the richest and poorest zip codes and compared the number of noise complaints per capita from each group over time.
There is a general increase in the number of complaints from both rich and poor zip codes over time, which was consistent with other analyses. At first, the richest and poorest zip codes had similar noise complaints per capita. Then, from 2013 to 2016, the richest neighborhoods had more noise complaints per capita than the poorest neighborhoods. However, something changed in 2017 and 2018, causing the poorest zip codes to have more noise complaints than the richest zip codes. More research is needed to discover the cause of this change and if it will stay the same in future years.
As highlighted in the correlation graph and map, on first inspection there was a positive correlation between the number of people under age 18 and the number of noise complaints in a given zip code each year. However, it looked as though the opposite became true once you started controlling for population within the variables. When we examined the relationship between the percent of the population under age 18 and the number of noise complaints per capita, we found that increasing the percent of kids in a zip code correlates with a decrease in the number of noise complaints per capita.
Another factor contributing to New York City’s reputation is the nightlife, of which live music is a contributor. Perhaps music and nightlife will be reflected in 311 complaints.
The choropleth map displays the number of music venues in each ZCTA. (The music venue data obtained from Ticketmaster’s Discovery API.) It is pretty clear that the ZCTAs with the highest concentration of venues are in Manhattan, from Midtown and south, and in ZCTAs in Brooklyn and Queens along the East River. Over 50% of the ZCTAs do not have a venue at all. The map makes it clear that venues are lacking in eastern Queens and the Bronx, southern Brooklyn and Staten Island. This is not too surprising as these areas tend to be residential or industrial.
The red dots cover areas that are approximately 1000 foot in radius from five established music venues in New York City. These venues are Madison Square Garden, the Apollo Theater, Terminal 5, the Bowery Ballroom, and the Kings Theatre. Clicking the red dots provide summary information about the venue, including capacity and number of 311 noise complaints within 1000 feet of the venue between 2012 and 2018.
Click on the buttons on the left to zoom to each venue. Click on the points around the venues to explore the associated Complaint Descriptor, Type and Date
The plots below show by descriptor, the aggregate noise complaints citywide and within 1,000 feet of the select music venues: Madison Square Garden, Apollo Theater, Terminal 5, Bowery Ballroom and the Kings Theatre. The plots easily show some similarities and differences in complaint descriptor frequencies as well as counts. Like Citywide trends, “Loud Music/Party” related noise complaints are the most frequent complaints near by the Apollo Theatre, Bowery Ballroom, and Kings Theatre. Interestingly, “Banging/Pounding” complaints were most frequent near Madison Square Garden, and “Construction” related complaints were most frequent near Terminal 5. Citywide, “Banging/Pounding” was the second most numerous complaint descriptor and “Construction” was the fourth most numerous complaint descriptor.
The summary table of Census Variables (“Children under 18”, “Housing Units”, “Median age”, “Median household income”, “Population”, “Population with bachelor’s degree or higher”, “Population % with a Bachelor’s Degree”, “Population % Children under 18”) provides another opportunity to explore noise complaints, both near music venues and citywide. The table allows for sorting, filtering, querying, and adjusting the number of rows displayed to enable an easier analysis.
The citywide correlations indicated that ZCTAs with more educated populations, on average yields generated more noise complaints. We sought to assess if the citywide associations between Census variables and noise complaints were consistent near the select music venues. Filtering the table to display census variable revealed the populations near Madison Square Garden and Terminal 5 were approximately twice as likely to have a Bachelor’s degree than the Citywide average, and the Apollo Theatre, and almost four times as likely to have a Bachelor’s degree compared to the population near Kings Theatre. Interestingly, the nearby areas surrounding of Madison Square Garden and Terminal 5 had the fewest noise complaints compared to other venues. Are these noise complaints totals and descriptors best explained by education levels?
The median income value for venue areas was the Bowery Ballroom, which had the most complaints. The areas near Terminal 5 and Madison Square Garden had the first and second highest median income values, and appeared to follow a different trend than we saw citywide. Could these be related to the type of complaints? Are wealthy people more likely to build or undertake construction projects?
We were curious to see how wealth and kids were associated with noise complaints. Would wealthy people complain more about noise? Or would they have fewer complaints, possibly reflecting a more quiet lifestyle? Would the presence of people under 18 be associated with increased noise complaints (possibly by parents seeking quiet for their families or by neighbors of families with noisy kids)? The interactive and user friendly table below allows one to easily emplore these questions and more.
Regarding the relationship between youth and complaints, there were about 20 to 50% fewer kids in ZIP Codes with music venues than in the citywide ZIP Codes. Looking at the table and bar charts did not allow for a clear association between percentage of the population under 18 and noise complaints. Perhaps these are more other omitted variables better suited to answer this question?
The below plots and data enable the user to explore questions related to Census variable and noise complaints. Please help yourself to generating your own hypotheses and explore the noise!
This project sought to highlight the patterns in 311 noise complaint data in New York City. We explored the connections between music venue locations, population attributes, time, and noise complaints. Recommendations for further research include understanding more about cause and effect: Are certain people more likely to make noise complaints than others? Or are there certain people, locations, and behaviors that are more likely to cause complaints than others? Do the attributes of a certain zip code attract people who are more or less likely to make noise complaints? Or do they adjust their behavior to the pre-established norms of a neighborhood? Have people become more familiar with 311 over time, and thus become more likely to call in complaints? Or have the noise levels in NYC truly increased over time?
You can find our process book, showcasing some decisions leading up to the creation of this project, HERE
NYC Open Data, 311 Service Requests from 2010 to Present, April 2019.
U.S. Census Bureau, 2010-2017 American Community Survey 5-Year Estimates, generated April 2020.
U.S. Census Bureau, TIGER/Line Shapefile, U.S., 2010 Census 5-Digit ZIP Code Tabulation Area (ZCTA5), 2018.
Ticketmaster, Discovery API v2, April 2019.
Jane Mills: jam2269@columbia.edu
William Peterson: wbp2114@columbia.edu
Gretchen Streett: gis2111@columbia.edu